Carcinogenicity studies require positive identification of test animals. Due to unreliability of ear notches and tags and inability to tattoo pigmented rodents, it is necessary to investigate more dependable and aesthetically acceptable identification methods that can be read directly or by electronic means. The purpose of this study is to determine the stability, readability and tissue reaction of a microchip glass sealed device when implanted in the subcutaneous tissue of B6C3F1 mice for two years. Seventy B3C3F1 mice/sex were anesthetized, implanted and housed individually in polycarbonate cages. The devices were read by a radio frequency scanner weekly and palpated at monthly intervals. Ten mice/sex were necropsied at 3 months and this will be repeated at 15 months with the remaining animals to be evaluated at 24 months. Two of the 140 devices were lost and 1 failed by day 10. Devices were palpable and appear to be fixed at one location with no inflammation or palpable masses at 8 months. At the 3 month necropsy, implants were encapsulated by thin fibrous tissue, and a clear cyst was present around one implant. The transponder- detector system is working satisfactorily. However, alternately sized scanning/reading systems may be required based on various animal housing requirements. Procedures are in progress to convert the 10 digit random alpha-numeric identification of the implant to a more practical user number sequence.